An efficient edge-preserving approach based on adaptive fuzzy switching median filter

This paper proposes a two-stage approach for detecting and removing of salt and pepper noise. In the first stage, the maximum and minimum in the contaminated image is used to identify noise pixels and noise-free pixels. In the second stage, these detected noise candidates will be restored by an improved adaptive fuzzy switching median algorithm, while these noise-free pixels are left unchanged. In term of the conflict of noise suppression and edge-preserving, our proposed method takes fuzzy switching reasoning to handle it. Simulation results show that it can remove salt and pepper noise effectively with edge-preserving compared to those nonlinear filters existing in the literature.